DEVELOPMENTAL HISTOET OF THE MOLLTJSCA. 
9 
ensuing growth is that the embryo cannot be got to tilt over on its side as in the earlier 
state. It persistently presents a strictly dorsal, haemal, or umbonal aspect, or an equally 
symmetrical, ventral, oral, or pedal aspect. 
Fig. 37 represents an embryo which is nearing this change, but has yet to develop a 
great margin of mantle-flap, which is the efficient cause of this change of habitual 
attitude. In the later stage the observer is at first puzzled as to what has become of 
the shell-gland and its groove. In fig. 37 some indication is afforded as to what develop- 
ment it is undergoing ( sh ). It becomes very much less obvious than hitherto, owing to 
the relative development of other parts and to the flattening of the arched surface on 
which it formerly sat as a saddle. Its area is at the same time very greatly increasing — 
that is to say, the cells all round the original oval patch of columnar epidermal cells are 
acquiring the same character, and ultimately the cells of the general surface of the 
mantle will assume the same character. The very large growth of the gastric cavity 
at this stage is remarkable. The whole embryo is of course now continually increasing 
in size, which is not indicated in the figures ; but the gastric chamber or pair of chambers 
have dilated into one great bilobed sac, to which the rectal peduncle forms but a small 
appendage. Up to this period the cellular elements of the walls of the gastric chamber 
have not presented any noticeable feature, ciliated on their inner surface, and apparently 
consisting of but a single series (though possibly a second series maybe present but not 
obvious) of corpuscular elements. In the stage to which we are about to pass, they 
appear to take on the most extraordinary activity ; and it becomes quite clear that what 
has up to this point functioned as an aproctous alimentary canal is a mere larval affair, 
and not even the rudiment of a part of the permanent digestive chamber. The cells or 
corpuscles of its walls proliferate and arrange themselves in new masses to form the 
permanent alimentary tract, and its glandular appendage the liver. The pharynx and 
the rectal peduncle are, however, unaffected by this process of re-formation. 
In Plate 3. fig. 37 the blind termination of the rectum is clearly seen. 
In Plate 3. fig. 38 it is again obvious ( rc ) ; and in this figure the first rudiments of 
the shell-valves, which now become evident, are introduced. The lower of the two is 
seen lying in contact with a part of the shell-gland (sh), which is in optical section, and 
extends really across the whole area occupied by the two shells. 
Plate 4. fig. 39 gives a dorsal or umbonal view of an embryo in the next stage of 
development — that in which the mantle has freely developed its large border. The 
length of the foot has now greatly increased, and the shell-valves are larger and more 
nearly approximating at their umbones than in figs. 38 & 38 a. Showing through 
the mantle-surface beneath the shells are the two lobes of the gastric chamber, now 
undergoing those curious developments of its cell-elements of which mention has just 
been made. 
Plate 4. figs. 40, 41, 42 represent a series of the modifications which the cell- 
elements of the gastric chamber undergo. They are taken from different embryos of 
three successive ages. A number of large pellucid nuclei first make their appearance, 
MDCCCLXXV. c 
